组合导航及其滤波算法研究.pdf

摘要
组合导航系统目前在人类的航空、航天等领域被广泛运用。本文对由惯性导
航系统（INS）和全球卫星定位系统（GPS）构成的组合导航系统进行了研究。
本文对惯性导航系统和全球卫星定位系统分别进行了详细的探讨，论述了二
者各自的工作原理、系统组成、误差来源和模型计算。计算机技术的迅猛发展促
进了卡尔曼（Kalman）滤波技术的发展，卡尔曼滤波技术的发展对组合导航系统
的发展有着重要的意义。常规卡尔曼滤波技术在实际应用中对系统的运动模型要
求较高，而这一要求一般很难得到保证，从而使得滤波不能得到系统的最优估计，
甚至可能引起滤波发散。
本文针对常规卡尔曼滤波在实际应用中存在的不足，论述了其他一些改进的
滤波技术：扩展卡尔曼滤波、无迹卡尔曼滤波、Sage-Husa 自适应滤波、 H ∞ 鲁棒
滤波。文中，对扩展卡尔曼滤波和无迹卡尔曼滤波在非线性情况下的表现进行了
简单仿真。本文对 INS/GPS 组合导航系统，引入了已知的误差模型，利用位置/速
度组合模式依据不同的滤波算法进行仿真实验。仿真结果表明对于 INS/GPS 组合
导航系统，改进的滤波算法对于滤波的精度和可靠性方面相比常规卡尔曼滤波表
现更为出色。


关键词：组合导航系统 卡尔曼滤波 INS/GPS 鲁棒滤波
Abstract
Integrated navigation system has been widely used in the human aviation,
aerospace and other areas currently. The study on the integrated navigation System
based on Inertial Navigation System (INS) and Global Positioning System (GPS) is
done in this paper.
In the paper, inertial navigation systems and global satellite positioning systems are
explored in detail, their working principle, system components, error sources and model
calculations are both discussed here. The rapid development of computer technology
has promoted the development of Kalman filtering technique and the development of
Kalman filtering has an important meaning to the development of integrated
navigation system. The require of the kinematic model of the system is higher in the
practical application of conventional Kalman filtering technique, this requirement is
difficult to be guaranteed in general, making the system can not be the optimal filter
estimate may even lead to filter divergence.
In this paper, conventional Kalman filter in practical applications its shortcomings,
as discussed in a number of other improved filtering techniques such as Extended
Kalman filtering, Unscented Kalman filtering, Sage-Husa adaptive filtering, robust
filtering. A simple simulation of the performance of the extended Kalm